This thesis reviews the foundational work on net neutrality and related concepts of free culture, the end to end principle, the neutral platform, and zero rating.
In 2000 Lessig and Lemley appropriated the end to end argument, an engineering concept and applied it to Internet and network access regulation in “End of End-to-End: Preserving the Architecture of the Internet in the Broadband Era.”15 Insisting that intelligence lies in the ends network, not the core, they declared that this end-to-end principle explains not only the internet’s operation and commercial success, but a justification for telecom regulation that will ensure the Internet’s continued functioning in the future. They equated this end to end concept with the notion of an “open” network to be facilitated by by
14 Free Press, “Net Neutrality: What You Need to Know Now,” Free Press, accessed June 20, 2016, http://www.savetheinternet.com/net-neutrality-what-you-need-know-now.
15 Mark A. Lemley and Lawrence Lessig, “The End of End-to-End: Preserving the Architecture of the Internet in the Broadband Era,” SSRN Electronic Journal, 2000, doi:10.2139/ssrn.247737.
“open access” policies. Such a regime is important to achieve the “free culture” digital commons which Lessig advocated in his subsequent book Free Culture: How Big Media Uses Technology and the Law to Lock Down Culture and Control Creativity.16
Tim Wu, who studied with Lessig, developed the term network neutrality.17 He too was concerned about intellectual property and “big business” locking down culture and innovation. His 2003 article includes a survey of the “discriminatory” practices and contracts of a number of fixed line ISPs. Though he posits that the interests of ISPs are inherently in cross purposes to the public, he does not present a structural model or theory of why ISPs would discriminate against third party services that use networks. Wu posits that rules may be necessary to restrict ISP behavior so that third party applications can enjoy “Darwinian” competition and end users can get the “best”
innovation.
Barbara van Schewick also studied with Lessig and published Internet Architecture and Innovation18 in 2010 which formalized the notion of
“end to end” as policy principle. She asserts that the end to end principle explains the “neutral” architecture of the Internet and that the Internet was expressly designed for the the benefit of the application at end points of the network. Van Schewick confirms that is is appropriate therefore to “suppress”19 ISP innovation in networks, traffic management and monetization, to favor a regime for innovation in third party applications.
Following these key arguments, the idea of a net neutrality policy is that it should control ISPs so that they do not get in the way of vital
16 Lawrence Lessig, “Free Culture,” Freeculture.pdf, 2004, http://www.free-culture.cc/freeculture.pdf.
17 Tim Wu, “Network Neutrality, Broadband Discrimination,” SSRN Scholarly Paper (Rochester, NY: Social Science Research Network, June 5, 2003), http://papers.ssrn.com/abstract=388863.
18 Barbara van Schewick, Internet Architecture and Innovation, New edition (Cambridge, MA: The MIT Press, 2010).
19 Personal interview. Barbara van Schewick. 31 August 2016.
end point applications nor restrict users who connect to the Internet expressly for these end point applications. The ISP should thus be neutral to all applications and operate the network without any
“discrimination” to applications. Net neutrality rules are thus the embodiment of preserving the Internet in this “neutral” and “open”
state for the primary benefit of users and third party innovation at end points. The needs of the network providers, whether their own innovation, efficiency, or solvency, are secondary.
There is some degree of interpretation between the academic assertions of net neutrality and the codification into rules. A review of the rules around the world finds that countries differ somewhat in their definitions, instruments, provisions, and punishments for the policy.
While the very activity of companies on the Internet suggests innovative forces at work, it is not clear to what degree “neutrality” or the regulated or legislated “Open Internet” is responsible for such innovation. However the verve and speed of the policymaking over the last decade suggests that policymakers’ should have an intuition about how the policy will work. Indeed if neutrality is necessary and desirable, there should be some idea about the optimal distribution of internet traffic and usage. If not, it will be difficult to tell whether the policy is moving the nation in the right direction.
It could be predicted that under neutrality that Internet traffic should lead to a random distribution to all possible end points, perhaps a Guassian (or bell curve) of distribution across all content and applications. Alternatively, one might imagine a a Pareto distribution of traffic, in which 80 percent of the traffic goes to 20 percent of the content and applications. However Wu’s notion of an evolutionary, meritocratic, “survival of the fittest” competition between all applications, suggests traffic should follow to “the very best innovation.” Perhaps today’s status quo is the manifestation of Wu’s prescription, as 99 percent of users go to 1 percent of the
destinations.20 Ostensibly this 1 percent is the “best” of the Internet’s innovation, following Wu’s definition.
Wu’s description of an neutral platform suggests there is perfect competition for content and applications, that any end point can compete equally with any other. In a perfectly competitive world, there are homogenous products, homogeneous users, perfect information, no barriers to entry, and no switching costs. However the fact that global internet traffic is highly concentrated to a few particular end points suggests that the market for apps is imperfectly competitive. Many of the top destinations on the Internet have held their position for more than a decade.
For example, the top most visited internet sites in the USA are each at least 10 years old.21 It seems counterintuitive that under a neutral regime that the rank of the most popular destinations would have changed so little over time. Indeed it would seem that “neutrality”
would enable more disruption to the established internet companies, that disruptors would more easily take the top positions in traffic, and that the “next Google” would emerge from those countries with the hardest net neutrality rules. In fact Google, which is almost 20 years old, has retained its top position as one of the most visited sites (if not the most visited site) in almost every country in the world in spite of the increasing number of countries with Open Internet rules.
The question is whether this traffic distribution is result of
“neutrality”, a competitive meritocracy in that users find these end points to be “the very best innovation” as Wu says. Could neutrality work in the opposite way, for example by cementing the position of the largest players to the detriment of new entrants? Could it be that rules that require treating data the same have the perverse effect of rewarding the large companies which already have revenue and users while harming the upstarts which need differentiation in order to be
20 Ramos, Andreas. “Can We Just Build It and They Will Come?,” Andreas.com, accessed October 31, 2016, http://andreas.com/can-we-just-build-it-and-they-will-come/.
21 “Top Sites in United States,” Alexa.com, accessed October 31, 2016, http://www.alexa.com/topsites/countries/US.
noticed? The question is then to what degree net neutrality rules create this “neutral platform” for “the very best innovation” on mobile networks and to what degree this model also works for startup innovation.
Such brisk rulemaking across so many countries would seem to indicate that policymakers believe net neutrality rules to be beneficial for their nations. It is surprising, however, that there is not more empirical research. Given the urgency and necessity of net neutrality rules, as it were, empirical research demonstrating the value of the policy would be more forthcoming. In addition one would expect that the national innovation policy authorities would have weighed in in support for such policies, given net neutrality’s purported link to innovation. But strangely, these innovation institutes have little to say on the topic.
In the US, the National Academies of Science, Engineering and Medicine organized a committee on Comparative National Innovation Strategies: Best Practice for the 21st Century.22 The committee met over 6 years and produced a number of reports including Innovation Policies for the 21st Century.23 It emphasizes the need to foster ecosystems, particularly local and regional ecosystems, as well as public-private partnerships. A search of the institute’s archives found very little information on the topic of the Internet and nothing on “net neutrality” or “open internet”.
The European Union also has a major research function in the European Commission’s Research and Innovation arm24 which by law must conduct research policies and implement research programs. It has not conducted research on net neutrality that is findable in its
22 “Comparative National Innovation Strategies: Best Practice for the 21st Century,”
The National Academies of Sciences, Engineering and Medicine, accessed October 31, 2016,
http://sites.nationalacademies.org/PGA/step/ComparativeInnovationPolicy/.
23 Innovation Policies for the 21st Century: Report of a Symposium (Washington, D.C.: National Academies Press, 2007), http://www.nap.edu/catalog/11852.
24 “Why European Research | Research & Innovation,” accessed November 9, 2016, http://ec.europa.eu/research/index.cfm?pg=why.
database, but its report on innovation “Models of Innovation in Global ICT Firms: The Emerging Global Innovation Ecosystems” see innovation occurring within an ecosystem as the result of the
“symbiotic” interplay of actors, including models in which telecom operators partner with different actors in the ecosystem.25
It appears that the “ecosystem” view of innovation in which actors work together symbiotically is the antithesis to the “net neutrality”
view in which one player needs to be controlled. While thousands of articles have described and debated net neutrality, almost none have tested it empirically within the context of national policymaking. This thesis tests the theory on apps in mobile networks and thus offers an important addition to address the gap in the literature.